Skip to content
Get packs

Audio Cleanup Editor

Removes filler words, stutters, dead air and mouth clicks from recordings, then re-levels them.

Source Download skill
Repository
danielmiessler/LifeOS
Stars
19,682
Licence
MIT
Files
15 files

What it does

When installed, the agent can take a spoken recording, transcribe it with Whisper, ask Claude to label each segment as keep or cut (fillers, false starts, stutters, dead air), and then cut those parts with ffmpeg, smoothing each join with a short crossfade. It also has separate tools to soften breaths and mouth clicks, check and fix 'ticking' caused by noise gates, and set loudness to a YouTube-style standard. Previews modify nothing, and cloud polishing runs only when requested.

Great for

  • Podcasters or narrators who want a fast first pass at removing ums, stutters and long silences
  • People fixing clicks or ticks in a recording caused by a noise gate or Voice Isolation
  • Users who already run LifeOS and want a repeatable audio cleanup workflow

Skip it if

  • Video editing or video composition, which this skill explicitly excludes (it defers to a separate video skill)
  • Publication-ready tight cutting where a human ear is needed, since the skill itself says meter-clean is not ear-clean and results must be listened to
  • Sensitive recordings you do not want sent to a third party, because the optional Cleanvoice polish step uploads audio to an external service

Try asking

  • “Clean the filler words and dead air out of this podcast episode and give me a preview first”
  • “Check my recording for gate ticking and repair it if there are defects”
  • “Tame the loud breaths and mouth clicks in this interview without touching the rest of the audio”
  • “Set this export to -14 LUFS for YouTube delivery”
Step-by-step guide: how to use Audio Cleanup EditorInstall, first run, examples, tips and FAQs

Install in 3 steps

  1. 1

    Download the skill

    The ZIP contains the 15 original files, the MIT licence and an attribution note.

    AudioEditor.zip
  2. 2

    Add it to your agent

    Claude Code (all projects)

    mkdir -p ~/.claude/skills && unzip AudioEditor.zip -d ~/.claude/skills/

    Claude Code / Agent SDK (this project only)

    mkdir -p .claude/skills && unzip AudioEditor.zip -d .claude/skills/

    Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.

  3. 3

    Ask for the task

    No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.

Get better results

  • Always run the preview mode or GateScan first so you can hear what will change before committing to edits.
  • Keep the original recording; the tools write new files such as _edited outputs, but listening to the result is the only real quality check.
  • Skip the Cleanvoice polish step for confidential audio unless you accept that it is uploaded to an external service.

What's inside

  • Intents/Clean.md
  • Setup.md
  • SKILL.md
  • Tools/Analyze.help.md
  • Tools/Analyze.ts
  • Tools/BreathTame.help.md
  • Tools/BreathTame.ts
  • Tools/Edit.help.md
  • Tools/Edit.ts
  • Tools/GateRepair.help.md
  • Tools/GateRepair.ts
  • Tools/GateScan.help.md
  • Tools/GateScan.ts
  • Tools/LoudnessLock.help.md
  • Tools/LoudnessLock.ts

Redistributed unmodified from danielmiessler/LifeOS under the MIT licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.